Assignments & Case Studies from the IIIT PG Diploma in Data Science (May 2020 - Dec 2021)

HELP International is an international humanitarian NGO that is committed to fighting poverty and providing the people of backward countries with basic amenities and relief during the time of disasters and natural calamities. It runs a lot of operational projects from time to time along with advocacy drives to raise awareness as well as for funding purposes. After the recent funding programmes, they have been able to raise around $ 10 million. Now the CEO of the NGO needs to decide how to use this money strategically and effectively. The significant issues that come while making this decision are mostly related to choosing the countries that are in the direst need of aid.

And this is where you come in as a data analyst. Your job is to categorise the countries using some socio-economic and health factors that determine the overall development of the country. Then you need to suggest the countries which the CEO needs to focus on the most. The datasets containing those socio-economic factors and the corresponding data dictionary are provided below.

Countries' Data

Data Dictionary

[Note - The inflation definition in the data dictionary has a typo. The correct definition would be the measurement of the annual growth rate of the GDP deflator]

Your main task is to cluster the countries by the factors mentioned above and then present your solution and recommendations to the CEO using a PPT. The following approach is suggested :

1. Start off with the necessary data inspection and EDA tasks suitable for this dataset - data cleaning, univariate analysis, bivariate analysis etc.
2. Outlier Analysis: You must perform the Outlier Analysis on the dataset. However, you do have the flexibility of not removing the outliers if it suits the business needs or a lot of countries are getting removed. Hence, all you need to do is find the outliers in the dataset, and then choose whether to keep them or remove them depending on the results you get.
3. Try both K-means and Hierarchical clustering(both single and complete linkage) on this dataset to create the clusters. [Note that both the methods may not produce identical results and you might have to choose one of them for the final list of countries.]
4. Analyse the clusters and identify the ones which are in dire need of aid. You can analyse the clusters by comparing how these three variables - [gdpp, child_mort and income] vary for each cluster of countries to recognise and differentiate the clusters of developed countries from the clusters of under-developed countries.
5. Also, you need to perform visualisations on the clusters that have been formed. You can do this by choosing any two of the three variables mentioned above on the X-Y axes and plotting a scatter plot of all the countries and differentiating the clusters. Make sure you create visualisations for all the three pairs. You can also choose other types of plots like boxplots, etc.
6. Both K-means and Hierarchical may give different results because of previous analysis (whether you chose to keep or remove the outliers, how many clusters you chose, etc.) Hence, there might be some subjectivity in the final number of countries that you think should be reported back to the CEO since they depend upon the preceding analysis as well. Here, make sure that you report back at least 5 countries which are in direst need of aid from the analysis work that you perform.

A well-commented Jupyter notebook containing the Clustering Models(both K-means and Hierarchical Clustering) and the final list of countries. Present the overall approach of the analysis in a presentation Mention the problem statement and the analysis approach. Explain the results of Clustering Model briefly. Include visualisations and summarise the most important results in the presentation. Make sure that you mention the final list of countries here ( Don't just mention the cluster id or cluster name here. Mention the names of all the countries.)

You need to submit the following two components

1. Python notebook: Should include detailed comments and should not contain unnecessary pieces of code
2. PPT: Make a PPT to present your analysis to the CEO (and thus you should include both the technical and the business aspects). The PPT should be concise, clear, and to the point. Submit the PPT after converting into the PDF format. The visualisations mentioned above must be present in this file.